1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/i18n/scriptset.cpp Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,276 @@
1.4 +/*
1.5 +**********************************************************************
1.6 +* Copyright (C) 2013, International Business Machines
1.7 +* Corporation and others. All Rights Reserved.
1.8 +**********************************************************************
1.9 +*
1.10 +* scriptset.cpp
1.11 +*
1.12 +* created on: 2013 Jan 7
1.13 +* created by: Andy Heninger
1.14 +*/
1.15 +
1.16 +#include "unicode/utypes.h"
1.17 +
1.18 +#include "unicode/uchar.h"
1.19 +#include "unicode/unistr.h"
1.20 +
1.21 +#include "scriptset.h"
1.22 +#include "uassert.h"
1.23 +
1.24 +U_NAMESPACE_BEGIN
1.25 +
1.26 +#define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
1.27 +
1.28 +//----------------------------------------------------------------------------
1.29 +//
1.30 +// ScriptSet implementation
1.31 +//
1.32 +//----------------------------------------------------------------------------
1.33 +ScriptSet::ScriptSet() {
1.34 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.35 + bits[i] = 0;
1.36 + }
1.37 +}
1.38 +
1.39 +ScriptSet::~ScriptSet() {
1.40 +}
1.41 +
1.42 +ScriptSet::ScriptSet(const ScriptSet &other) {
1.43 + *this = other;
1.44 +}
1.45 +
1.46 +
1.47 +ScriptSet & ScriptSet::operator =(const ScriptSet &other) {
1.48 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.49 + bits[i] = other.bits[i];
1.50 + }
1.51 + return *this;
1.52 +}
1.53 +
1.54 +
1.55 +UBool ScriptSet::operator == (const ScriptSet &other) const {
1.56 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.57 + if (bits[i] != other.bits[i]) {
1.58 + return FALSE;
1.59 + }
1.60 + }
1.61 + return TRUE;
1.62 +}
1.63 +
1.64 +UBool ScriptSet::test(UScriptCode script, UErrorCode &status) const {
1.65 + if (U_FAILURE(status)) {
1.66 + return FALSE;
1.67 + }
1.68 + if (script < 0 || script >= (int32_t)sizeof(bits) * 8) {
1.69 + status = U_ILLEGAL_ARGUMENT_ERROR;
1.70 + return FALSE;
1.71 + }
1.72 + uint32_t index = script / 32;
1.73 + uint32_t bit = 1 << (script & 31);
1.74 + return ((bits[index] & bit) != 0);
1.75 +}
1.76 +
1.77 +
1.78 +ScriptSet &ScriptSet::set(UScriptCode script, UErrorCode &status) {
1.79 + if (U_FAILURE(status)) {
1.80 + return *this;
1.81 + }
1.82 + if (script < 0 || script >= (int32_t)sizeof(bits) * 8) {
1.83 + status = U_ILLEGAL_ARGUMENT_ERROR;
1.84 + return *this;
1.85 + }
1.86 + uint32_t index = script / 32;
1.87 + uint32_t bit = 1 << (script & 31);
1.88 + bits[index] |= bit;
1.89 + return *this;
1.90 +}
1.91 +
1.92 +ScriptSet &ScriptSet::reset(UScriptCode script, UErrorCode &status) {
1.93 + if (U_FAILURE(status)) {
1.94 + return *this;
1.95 + }
1.96 + if (script < 0 || script >= (int32_t)sizeof(bits) * 8) {
1.97 + status = U_ILLEGAL_ARGUMENT_ERROR;
1.98 + return *this;
1.99 + }
1.100 + uint32_t index = script / 32;
1.101 + uint32_t bit = 1 << (script & 31);
1.102 + bits[index] &= ~bit;
1.103 + return *this;
1.104 +}
1.105 +
1.106 +
1.107 +
1.108 +ScriptSet &ScriptSet::Union(const ScriptSet &other) {
1.109 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.110 + bits[i] |= other.bits[i];
1.111 + }
1.112 + return *this;
1.113 +}
1.114 +
1.115 +ScriptSet &ScriptSet::intersect(const ScriptSet &other) {
1.116 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.117 + bits[i] &= other.bits[i];
1.118 + }
1.119 + return *this;
1.120 +}
1.121 +
1.122 +ScriptSet &ScriptSet::intersect(UScriptCode script, UErrorCode &status) {
1.123 + ScriptSet t;
1.124 + t.set(script, status);
1.125 + if (U_SUCCESS(status)) {
1.126 + this->intersect(t);
1.127 + }
1.128 + return *this;
1.129 +}
1.130 +
1.131 +UBool ScriptSet::intersects(const ScriptSet &other) const {
1.132 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.133 + if ((bits[i] & other.bits[i]) != 0) {
1.134 + return true;
1.135 + }
1.136 + }
1.137 + return false;
1.138 +}
1.139 +
1.140 +UBool ScriptSet::contains(const ScriptSet &other) const {
1.141 + ScriptSet t(*this);
1.142 + t.intersect(other);
1.143 + return (t == other);
1.144 +}
1.145 +
1.146 +
1.147 +ScriptSet &ScriptSet::setAll() {
1.148 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.149 + bits[i] = 0xffffffffu;
1.150 + }
1.151 + return *this;
1.152 +}
1.153 +
1.154 +
1.155 +ScriptSet &ScriptSet::resetAll() {
1.156 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.157 + bits[i] = 0;
1.158 + }
1.159 + return *this;
1.160 +}
1.161 +
1.162 +int32_t ScriptSet::countMembers() const {
1.163 + // This bit counter is good for sparse numbers of '1's, which is
1.164 + // very much the case that we will usually have.
1.165 + int32_t count = 0;
1.166 + for (uint32_t i=0; i<LENGTHOF(bits); i++) {
1.167 + uint32_t x = bits[i];
1.168 + while (x > 0) {
1.169 + count++;
1.170 + x &= (x - 1); // and off the least significant one bit.
1.171 + }
1.172 + }
1.173 + return count;
1.174 +}
1.175 +
1.176 +int32_t ScriptSet::hashCode() const {
1.177 + int32_t hash = 0;
1.178 + for (int32_t i=0; i<LENGTHOF(bits); i++) {
1.179 + hash ^= bits[i];
1.180 + }
1.181 + return hash;
1.182 +}
1.183 +
1.184 +int32_t ScriptSet::nextSetBit(int32_t fromIndex) const {
1.185 + // TODO: Wants a better implementation.
1.186 + if (fromIndex < 0) {
1.187 + return -1;
1.188 + }
1.189 + UErrorCode status = U_ZERO_ERROR;
1.190 + for (int32_t scriptIndex = fromIndex; scriptIndex < (int32_t)sizeof(bits)*8; scriptIndex++) {
1.191 + if (test((UScriptCode)scriptIndex, status)) {
1.192 + return scriptIndex;
1.193 + }
1.194 + }
1.195 + return -1;
1.196 +}
1.197 +
1.198 +UnicodeString &ScriptSet::displayScripts(UnicodeString &dest) const {
1.199 + UBool firstTime = TRUE;
1.200 + for (int32_t i = nextSetBit(0); i >= 0; i = nextSetBit(i + 1)) {
1.201 + if (!firstTime) {
1.202 + dest.append((UChar)0x20);
1.203 + }
1.204 + firstTime = FALSE;
1.205 + const char *scriptName = uscript_getShortName((UScriptCode(i)));
1.206 + dest.append(UnicodeString(scriptName, -1, US_INV));
1.207 + }
1.208 + return dest;
1.209 +}
1.210 +
1.211 +ScriptSet &ScriptSet::parseScripts(const UnicodeString &scriptString, UErrorCode &status) {
1.212 + resetAll();
1.213 + if (U_FAILURE(status)) {
1.214 + return *this;
1.215 + }
1.216 + UnicodeString oneScriptName;
1.217 + for (int32_t i=0; i<scriptString.length();) {
1.218 + UChar32 c = scriptString.char32At(i);
1.219 + i = scriptString.moveIndex32(i, 1);
1.220 + if (!u_isUWhiteSpace(c)) {
1.221 + oneScriptName.append(c);
1.222 + if (i < scriptString.length()) {
1.223 + continue;
1.224 + }
1.225 + }
1.226 + if (oneScriptName.length() > 0) {
1.227 + char buf[40];
1.228 + oneScriptName.extract(0, oneScriptName.length(), buf, sizeof(buf)-1, US_INV);
1.229 + buf[sizeof(buf)-1] = 0;
1.230 + int32_t sc = u_getPropertyValueEnum(UCHAR_SCRIPT, buf);
1.231 + if (sc == UCHAR_INVALID_CODE) {
1.232 + status = U_ILLEGAL_ARGUMENT_ERROR;
1.233 + } else {
1.234 + this->set((UScriptCode)sc, status);
1.235 + }
1.236 + if (U_FAILURE(status)) {
1.237 + return *this;
1.238 + }
1.239 + oneScriptName.remove();
1.240 + }
1.241 + }
1.242 + return *this;
1.243 +}
1.244 +
1.245 +U_NAMESPACE_END
1.246 +
1.247 +U_CAPI UBool U_EXPORT2
1.248 +uhash_equalsScriptSet(const UElement key1, const UElement key2) {
1.249 + icu::ScriptSet *s1 = static_cast<icu::ScriptSet *>(key1.pointer);
1.250 + icu::ScriptSet *s2 = static_cast<icu::ScriptSet *>(key2.pointer);
1.251 + return (*s1 == *s2);
1.252 +}
1.253 +
1.254 +U_CAPI int8_t U_EXPORT2
1.255 +uhash_compareScriptSet(UElement key0, UElement key1) {
1.256 + icu::ScriptSet *s0 = static_cast<icu::ScriptSet *>(key0.pointer);
1.257 + icu::ScriptSet *s1 = static_cast<icu::ScriptSet *>(key1.pointer);
1.258 + int32_t diff = s0->countMembers() - s1->countMembers();
1.259 + if (diff != 0) return diff;
1.260 + int32_t i0 = s0->nextSetBit(0);
1.261 + int32_t i1 = s1->nextSetBit(0);
1.262 + while ((diff = i0-i1) == 0 && i0 > 0) {
1.263 + i0 = s0->nextSetBit(i0+1);
1.264 + i1 = s1->nextSetBit(i1+1);
1.265 + }
1.266 + return (int8_t)diff;
1.267 +}
1.268 +
1.269 +U_CAPI int32_t U_EXPORT2
1.270 +uhash_hashScriptSet(const UElement key) {
1.271 + icu::ScriptSet *s = static_cast<icu::ScriptSet *>(key.pointer);
1.272 + return s->hashCode();
1.273 +}
1.274 +
1.275 +U_CAPI void U_EXPORT2
1.276 +uhash_deleteScriptSet(void *obj) {
1.277 + icu::ScriptSet *s = static_cast<icu::ScriptSet *>(obj);
1.278 + delete s;
1.279 +}
